Logo Lanfrica

Kiswahili Labelled Dataset

Domaine:

natural language processing

Type de record:

dataset
Créateur:
Kam
Éditeur:
Edi
Hôte:avatar
Data collected for a master's dissertation focusing on sentiment analysis for Kiswahili. Collected from online sources between 8.01.2025 and 31.01.25 in Nairobi, Kenya. Two datasets, one comprising 3,282 labelled Kiswahili text and one 3,117 labelled Kiswahili sentences. More information is included in the Readme document.